1

Abstract Configuration space computation is a

transformation process that reduces a robot to a single reference
point by expanding obstacles on the image plane. The obstacles
can be expanded by inverting the robot along a reference point
and then slide this reference point along their borders. The area
covered by the union of inverted robot during the sliding along
with the obstacles defines the configuration space of obstacles.
This approach reduces a complex problem into a simple one. In
this paper, we present a parallel algorithm for computing the
configuration space obstacles by using reconfigurable mesh
multiprocessors. The reconfigurable mesh multiprocessor system
is a multiprocessor model with flexible bus connection
capabilities. The digitized images of the obstacles and the robot
are stored in an image plane. The algorithm takes O(1) time and
is optimal.

Index Terms Configuration space, robotics, image
processing, parallel algorithms, Reconfigurable mesh
I. INTRODUCTION
CONFIGURATION space computation found applications in
motion planning, computer graphics, robot-assisted
surgery, automated assembly plans among many others.
For example, Wytyczak-Partyka et. al[15]. propose no fly zone
concept to assist surgeons. By defining the configuration space
of the instrument, their system can provide a collision free
working space for surgeons. In computer graphics application,
Bandi and Thalmann adopted Configuration space approach to
simulate human finger animation [1]. In [4], Ivanisevic and
Lumelsky used configuration space as means to enhance
human performance in teleoperation tasks. Because computing
configuration space concept provides a generalized framework
to study the motion planning problem and therefore is an
important problem in path planning for automatic robotics
applications see [3], [10], [11], [12], [13], [17].
Our aim in this paper is to develop constant time algorithm
for computing the configuration space on reconfigurable mesh
multiprocessors (RMESH). In [9], Kavraki used a Fast Fourier
Transform based algorithm to compute configuration space
obstacles. The objective of path planning is to find a path to

Manuscript revised June 30, 2011.
John Jenq is with the Department of Computer Science, Montclair State
University, Montclair NJ 07043 USA ( phone: 973-655-7237; fax: 973-655-
4164; e-mail: jenqj@ mail.montclair.edu).
Dajin Wang is with the Department of Computer Science, Montclair State
University, Montclair NJ 07043 USA (e-mail: wangd@mail.montclair.edu).
Wingning Li is with the Department of Computer Science and Computer
Engineering, University of Arkansas, Fayetteville AR 72701 USA (e-mail:
wingning@uark.edu).
move a robot A from a position s (the initial position) to
another position d (the final position) without colliding with
the obstacles already in space R. A common way to solve this
problem is the configuration space approach which reduces the
robot A to a single reference point p and expands each
obstacle
j
B

to include all the positions of p that cause a
collision between A and
j
B .The expansion of an obstacle
j
B
is called the configuration space obstacle of
j
B . In the new
representation, the object A (robot) becomes a single point.
The configuration space approach then effectively reduces a
complex problem to a simple one.


























To calculate the configuration space obstacle of an
obstacle
j
B , one can firstly invert robot A, i.e. to rotate A about
a reference point, say r, by 180 and then slide the reference
point around the boundary of obstacle
j
B . The union of the
areas covered by A during the sliding, and the area originally
covered by
j
B defines the configuration space obstacle of
j
B .
Figure 1 shows a robot A with reference point Ar and the
inverted robot with reference point Ar. Figure 1 also shows
the configuration space obstacle derived by using robot Ar and
Computing the Configuration Space on
Reconfiguration Mesh Multiprocessors
John Jenq, Dajin Wang, and Wingning Li
C
Ar
Ar
Fig. 1. Compute configuration space with robot inversion
Fig. 2. A point robot r and the expanded obstacles B
1
& B
2

B
1

B
2

r
Cyber Journals: Multidisciplinary Journals in Science and Technology, Journal of Selected Areas in Robotics and Control (JSRC), June Edition, 2011


2

its inversion Ar respectively. Figure 2 shows an example of
two obstacles
1
B and
2
B . The areas enclosed by the dark lines
are the configuration space obstacles of
1
B and
2
B . Note the
triangular robot A becomes a point r.

Parallel algorithms targeted at different architectures had
been proposed to speed up the whole process of path planning.
For example, Dehne, Hassenklover, and Sack have presented a
systolic algorithm for computing the configuration space
obstacles in a plane for a rectilinear convex robot [2]. Their
algorithm takes O(N) time for an N N image on an N N
mesh computer. Tzionas, Thanailakis, and Tsalides have
presented a parallel algorithm for collision free path planning
of a diamond-shaped robot and its implementation in VLSI
[16]. Jenq and Li developed optimal algorithms for computing
the configuration space for circular, rectangular and convex
robots by using hypercube computers [7], [8]. Their algorithms
run in O(logN) time for an N N image by using N N
processors and are optimal for hypercube computers.
In this paper, we consider convex robots and convex
obstacles. The digitized bitmap image of a convex robot is a
rectilinear convex polygon. Note the converse statement may
not be true. A polygon is rectilinear convex if (1) the polygon
is formed by horizontal and vertical line segments, and (2) the
intersection of the polygon with any horizontal or vertical line
consists of at most one line segment.

Since the class of reconfigurable mesh computers is a
superset of the class of mesh computers, the algorithm
developed by Dehne, Hassenklover, and Sack can be easily
simulated with the same complexity, i.e., O(N), on a RMESH.
In this paper, a constant time algorithm to compute
configuration space on an D N N RMESH is developed,
where D is the diameter of the robot. We can achieve same
time complexity and at the same time reduce the number of
processor to N N when the shape of the robot is either
rectangular or circular.
We organize the remainder of the paper as follows. In
section 2, we briefly describe the basic architecture and
configuration of RMESH. In section 3, we list and develop
some new fundamental RMESH data manipulation operations.
These operations are functioned as building blocks on which
the configuration space algorithms are developed. In section 4,
the constant time algorithm for computing configuration space
obstacles with a convex robot is discussed. We conclude this
report in section 5.
II. PRELIMINARIES ON RMESH
The particular reconfigurable mesh architecture that we use
in this paper is called RMESH[14]. It employs a
reconfigurable bus to connect together all processors. Figure 3
shows a 2 4 4 RMESH. By opening some of the switches,
the bus may be reconfigured into smaller buses that connect
only a subset of the processors. The flexible connection
capability makes RMESH a powerful model to generate
efficient solutions for various applications.



















The important features of an RMESH are:
1. An L M N RMESH is a 3-dimensional mesh-
connected array of processing elements (PEs).
Each PE in the RMESH is connected to a
broadcast bus, which is itself constructed as a
L M N grid. The PEs are connected to the
bus at the intersection of the grids. Each PE
manages up to six bus switches (see Fig. 3) that
are software controlled and can be used to
reconfigure the bus into sub buses. The ID of
each PE is a triple (i , j, k ) where i is the row
index, j is the column index and k is the plane
index. The ID of the upper left corner PE on
plane zero is (0,0,0) and that of the lower right
one is (N-1,M-1,0).
2. The six switches associated with each PE are
labeled as E (east), W (west), S (south), N
(north), B (back), and F (front). Notice that the
east (west, north, south, back, front) switch of a
PE is also the west (east, south, north, front,
back) switch of the PE (if any) on its right (left,
top, bottom, back, front). Two PEs can
simultaneously set (connect, close) or unset
(disconnect, open) a particular switch as long as
the settings do not conflict. The broadcast bus
can be subdivided into subbuses by opening
(disconnecting) some of the switches.
3. Only one of the processors connected to a given
subbus can broadcast its data on the subbus at
any time.
4. In unit time, data put on a subbus can be read by
every PE connected to it. Command broadcast(I)
is used by a PE to broadcast the value in its
register I to all the PEs on its subbus.
5. The statement R = content(bus) is used by a PE
to read the content of the bus into its R register.
6. Row buses are formed when each processor
disconnects (opens) its S switch, B switch, and
connects (closes) its E switch. The column buses
Switch Processor (PE) Link
Fig. 3. A 2 4 4 RMESH

3

can be formed by disconnecting the E and B
switches, and connecting the S switch of each
PE. Similarly, Z buses can be formed by
connecting F (or B) switch and disconnecting E
and S switches of each PE, while the plane buses
can be formed when each PE only disconnects its
B switch.

3.1. Broadcast
In a data broadcast operation, data originated in one PE are
sent to the remaining N -1 PEs, where N is the total number of
PEs in the RMESH network. This operation takes O(1) time.

3.2 Diagonalization
This operation will diagonalize a row (column) of elements,
by which we mean moving a specific row (column) elements to
diagonal positions with respect to that row (column). See
Figure 4 for illustration. With the RMESH bus, this operation
can be done in O(1) time.











3.3. Rank
Each PE(i) has a flag selected(i), which is set to true if PE(i)
is selected. A rank operation assigns a rank to each PE, where
the rank of PE(i), rank(i), is the number of selected PEs whose
indices are less than i. This operation takes O(logN) time.
However, N elements on a single row can be ranked in O(1)
time on an N N RMESH [6].

3.4. Shift
Each PE has data in its A variable that is to be shifted to B
variable of a processor that is s units, s > 0, to right or left in
the same row (column). A variant of shift is the operation of
circular shift, which performs shift with wrap-around. These
operations can be done in O(s) time. If 1 = s then the time
becomes O(1). However, shifting a row of m elements for
distance s can be done in O(1) time, if the s s m + ) (
neighboring PEs are available to use. The procedure is given
in Figure 5.

Step1 Partition the m elements into

s
m
blocks.
Step2 Diagonalize each of the s elements upward onto the
corresponding s s block.
Step3 Form row subbuses for diagonal elements between
evenodd blocks (i.e., block pairs ( 1 0 ),
( 3 2 ), , etc.)
Step4 PEs on even blocks broadcast(A), where A is the
value to be shifted.
Step5 PEs on diagonal of odd blocks do B=content(bus).
Step6 Column buses are formed on odd blocks.
Step7 PEs on diagonal of odd blocks broadcast(B)
Step8 The s elements on the bottom row of odd blocks do
B=content(bus)
Step9 (Phase 2) Repeat Step4 through Step8 for odd-even
blocks (i.e., block pairs ( 2 1 ), ( 4 3 ), ,
etc).

Fig. 5. Constant time algorithm for shift operation.

Figure 6 shows the two-phase shift operation for 20 = m
elements by using 4 24 PEs. The 20 PEs at bottom row are to
be shifted 4 positions to the left. The 4 24 PEs are
partitioned into six 4 4 blocks. The arrows represent data
movement. If wrapped around shift is required then extra steps
are needed to handle this. We omit the details here. The
complexity can be easily seen to be O(1).


























3.5. DrawSegment
This operation is defined only for PEs on the same row or
column, for simplicity, we will use just one index to identify a
processor, i.e., we use PE(i) to identify a processor in the
implied row or column under consideration. Each PE(i) has a
flag mark(i), a variable A(i), and another variable ext(i). PE(i)
is marked if mark(i) true. A DrawSegment operation transmits
the A(i) value of each marked PE( i ), to PE( i ), PE( i+1
),...,PE( i+ext(i) ) or stop propagating when the other PE
whose mark value true is encountered. This implementation
takes O(1) time as the following. Without loss generality, let
us assume A(i) = 1 are the same for all marked PEs.
Fig. 4. Diagonalization of a row of 4 elements
Block4 Block2 Block0

1
st
move

2
nd
move 3
rd
move
Block 1 Block 3
Block4 Block2 Block0

Block 1 Block 3
(a) Even phase shift
(b) Odd phase shift
Fig. 6. Two phase shifting: (a) 1
st
phase (b)2
nd
phase

4

Furthermore, let us assume we will draw segments for the
processors on the column 0 of plane 0 whose mark(i) is true
and toward south. The procedure is in Figure 7


Step1 if mark(i) then disconnect N and B switches
Step2 if mark(i) then broadcast(i)
Step3 index(k) = content(bus), for N k < s 0
Step4 if mark(i) then broadcast(ext(i))
Step5 ext(k) = content(bus) for N k < s 0
Step6 if ) ) ( ) ( ( k k ext k index > + and (mark(k) = false)
then {mark(k) = true; A[k] = 1}, for N k < s 0

Fig. 7. A constant time DrawSegment operation

Step1 form the column buses for the one dimensional
RMESH under consideration. Step2 through Step5 send the
row index of marked PE and its intended ext. value downward.
Step6 is to determine which PEs are inside the range of the ext.
of the marked PE above. For those PEs who are inside the
range of the ext. set their A values.

3.6. AdjacentUnion
This operation is similar to the DrawSegment operation
except that the A(i) is always of value 1. The other difference
is that when i+ext(i) of PE(i) is greater than j, for mark(j) =
true and j > i, the A[i] value(which is one) continue
propagating until PE with index i+ext(i) is encountered, while
in DrawSegment operation the propagating value A[i] stops
when PE( j) is encountered. This implementation takes
O(logN) time when there are N processors and can be done by
recursive doubling on the size of the column buses and update
the ext. values downward. It is similar to the hypercube
operation used in [5] to compute the area of MAT. Since we
are concerning constant time algorithms, there are two ways
one can do to reduce the complexity to O(1). Case (1) If the
extended lengths ext(i) are the same for the participant PEs,
and case (2) If there are at least ) (i ext N PEs available.
Let us examine these two methods separately. For case 1,
DrawSegment can be used to perform the task. The rational is
that during the drawing of the segment when a marked PE is
encountered the propagation stops. Fortunately, the uncovered
portion, that shall be drawn will be covered (drawn) by the
encountered PE (which will draw the same value). This is
exactly the dominate property mentioned in [7].
As for case 2, we assume there are )) ( max( i ext N PEs
available; where the max(ext(i)) is the diameter, D, of the
robot. Let us assume that all the N PEs participating in the
operation are in the same row. We firstly partition the N
processors into

D N / partitions. The algorithm will run
twice, one for the even blocks and the other for odd blocks.
Each time when a block is processed, two blocks of processors
are needed. Note each block is of size D D . The operation is
very similar to the constant time shift operation mentioned
earlier. The procedure is shown in Figure 8.


Step1 Diagonalization(ext(i))
Step2 DrawSegment(ext(i)) for PEs on the diagonal of the
block
Step3 The PEs that are drawn from Step2 form column bus
by disconnect N switch
Step4 Broadcast (1)
Step5 A[i](0,j)=content(bus)

Fig. 8. A constant time AdjacentUnion operation

Step1 transfers the ext values to the diagonal PEs by
Diagonalization operation. Step2 draws segment for each PE
on the diagonal line based on the ext value received from
Step1. At Step4, the PEs that are marked by the DrawSegment
operation broadcast the value one to the PEs at row 0. This can
be done by firstly setting up the column bus as in Step3. The
AdjacentUnion operation completes at Step5 when the A[i] is
received, if there is any. Figure 9 shows an example for this
operation. The numbers on the bottom represent ext. values.
The horizontal arrow lines are DrawSegment operation, while
the vertical arrow lines stands for broadcasting operation of
Step4. Note in the example, after the AdjacentUnion
operation, the PEs are all marked except the one that is in the
rightmost position.












3.7. Inversion
This operation rotates a rectilinear polygon by 180 around a
given valid reference point (i,j). A reference point (i,j) is valid
iff i, j are integers in the range of 0...N-1 and after the rotation
the rectilinear polygon remains within the N N image plane.
This operation can be accomplished in constant time on an
N N RMESH provided the gray value of each pixel in the
image of the rectilinear polygon is identical to one another.
The procedure is outlined in Figure 10.

Step1 Reference point broadcasts its i and j indices to all
pixels of the robot.
Step2 All right and left boundary pixels for the rectilinear
polygon identify themselves.
Step3 Every right boundary pixel collects length
information of its row segment and computes newJ
index after the inversion.
0 2 3 0 6 0 7
Ext values
Fig. 9. AdjacentUnion operation on RMESH

5

Step4 The PEs corresponding to the right boundary pixels
do diagonalization on window of H H with the
information of newJ and length information
calculated at Step3, where H is the height of the
rectilinear polygon.
Step5 Diagonal PEs broadcast newJ and length.
Step6 The PEs on the off-diagonal of the H H window
receive the information.
Step7 Setup row buses and broadcast newJ and length.
Step8 if received j newJ = index for PEs in the H H
window then do DrawSegment(length).
Fig. 10. Inversion operation for a rectilinear polygon.




































Figure 11 shows the inversion process of a simple rectilinear
polygon. Figure 12 illustrates Steps 3 through 8 of the
procedure.

In Step1, the reference point PE, using the plane bus,
broadcasts its coordinates (i,j) to all the PEs. Since the robot is
rectilinear convex, it can be decomposed into a set of unit
width horizontal segments. In Step2, after each PE checks its
right and left neighbors, the PEs located at either ends of the
horizontal segments can identify themselves. Once the left and
right boundary PEs have identified themselves, they can
determine the segment lengths for all the unit width horizontal
segments. The length computation is accomplished by first
setting up row buses, then each left boundary PE broadcasting
the column index of its left neighbor on its row bus, and finally
each right boundary PE receiving the index on its row bus and
subtracting it from its column index to get the segment length.
From laws of geometry, when a line segment is rotated
by 180 , the right end point of the line segment becomes the
left end point of the rotated line segment. Since line segments
are preserved under rotation, the length information of a line
segment would be sufficient to reconstruct the rotated line
segment if the coordinates of its left end point is known. The
coordinates of the left end point is determined by the PE,
currently located at the right end point of the corresponding
not yet rotated segment, by applying the transformation matrix
to its coordinates. This is done in Step3. After this step, the
right boundary PEs need to send their segment lengths to the
corresponding PEs located at the left boundary of the rotated
segments. This can be accomplished by diagonalization
operation, followed by column bus broadcast, and finally
followed by row bus broadcast. These are done in Steps 4 to 7.
Step8 reconstructs the polygon. The operations in each step
can be done in parallel. The time complexity is O(1).















III. COMPUTATION OF CONFIGURATION SPACE ON RMESH
Using the fundamental operations developed in the previous
section, we present a constant time RMESH algorithm to
compute configuration space obstacles for those robots of
which the digitized images may be modeled by WBPs (well
behaved polygons). Briefly speaking, a WBP is a polygon that
can be partitioned into at most four L-shaped polygons as
shown in Figure 13. The reader is referred to [8] for a more
detailed discussion of this type of polygons. Note that the
digitized images of commonly encountered robot shapes, such
as circles, rectangles, or convex polygons (possibly with
rotation), are WBPs. The intersection of the two dotted lines,
in Figure 13, is called the base point.

Some instances of the WBPs may have two base points. An
example of such an instance is shown in Figure 14. A
Base
point
Fig. 13. A WBP convex robot partitioned into four L-shaped polygons
length
newJ
Reference point
Fig. 11. Inversion of a rectilinear polygon
Step7 &
Step 8
Step5 &
Step6
Step4
NewJ = j

Right
boundary
DrawSegment
length
Fig. 12. Illustration of step3 to step8 of Inversion

6

technique of applying shift operations on the obstacles to
reduce WBPs having two base points to that having one base
point is developed in [8]. The same technique is used here.
The reader is referred to [8] again for an elaborated discussion












The computation of configuration space obstacles for a
WBP shaped robot is reduced to that for a L-shaped robot. The
final configuration space obstacles are computed by applying
at most four iterations of the algorithm, that computes the
configuration space obstacles for a L-shaped robot, and taking
the union of the configuration space obstacles obtained from
these iterations. Since the algorithms for the four different
kinds of L-shaped robots are basically symmetrical, we only
present the procedure for L-shaped robots having the base
point at their upper right corner. Such a L-shaped robot is
simply referred to as a robot in the remainder of the paper.
Before we proceed any further, let us note that the
information describing the robot is needed by each obstacle
PE, so that the PEs know how to expand the obstacles
simultaneously, as if each obstacle has the robot slid around its
boundary concurrently.
During the obstacle expansion, each obstacle first expands
itself vertically, and then horizontally. For vertical expansion,
each obstacle PE simply marks H PEs to its south as obstacle
PEs, where H is the height of the robot. Once each obstacle PE
receives the broadcasted H value, it can expands itself south-
ward in O(1) time by applying the DrawSegment operation of
Section 3.5.
Unlike vertical expansion, where before the expansion all
obstacle PEs are the original obstacle PEs, horizontal
expansion involves obstacle PEs that may be the original
obstacle PEs or the new obstacle PEs due to vertical
expansion. Hence, different horizontal expansion lengths may
be required by different obstacle PEs.
For an original obstacle PE, the length of 0
th
(top) horizontal
segment of the robot is used as its horizontal expansion length.
For a new obstacle PE, its expansion length depends on its
vertical distance from the original obstacle PE at the boundary.
For a new obstacle PE, if this distance is k, the length of k-1
th

horizontal segment of the robot is used for its horizontal
expansion. Thus, during horizontal expansion phase, each
obstacle PE not only needs to know its vertical distance from
the original obstacle PE (for an original obstacle PE this
distance is 0), but also needs to know its horizontal expansion
length.
In order to carry out the index (vertical distance) based
retrieval of the length information in constant time, the
following tiling procedure is developed. The procedure tiles
the length and index of each horizontal segment of the robot
for future reference, and operates on an D N N RMESH,
where D is the height of the robot, i.e., the number of unit
width horizontal segments that the robot has. The tiling
procedure is shown as in Figure 15.

Step1 Use shift operations to identify boundary pixel of
the L-shaped robot
Step2 If right boundary pixel then setup row bus by
disconnecting E, and B switches and broadcast its
j index on the bus
Step3 A= content(bus) for PE(i,0,0)
Step4 If left boundary pixel then form row bus by
disconnecting E, and B switches and broadcast its
j index on the bus
Step5 B=content(bus) for PE(i,0,0)
Step6 For PE(i,0,0) that receive A and B do runLength
:= A-B and form Z bus
Step7 Rank from top to down for PEs that received A
and B; put rank result in R
Step8 Broadcast(runLength) on Z bus for PE (i,0,0)
from Step6
Step9 runLength=content(bus) for PE(i,0,k), where
s k < s 0
Step10 Broadcast(R) on Z bus for PE (i,0,0) from Step6
Step11 D=content(bus) for PE(i,0,k), where s k < s 0
Step12 Form plane bus
Step13 Broadcast(runLength) for PE(i,0,k) and k==D
Step14 runLength=content(bus)


Fig. 15. Tiling of the length information for L-shaped polygon

Let l
0
, l
1
, l
2
, l
d-1
, be the lengths of the horizontal segments
of a robot from top to bottom respectively. Let rnuLength be a
register that each PE has. The goal of the tiling procedure is to
assign l
i
to all the runLength registers in plane i, 1 0 s s D i .
Of course, l
0
, l
1
, l
2
, l
d-1
, must be first calculated by the tiling
procedure, and then distributed to different planes.
Step1 uses four shift operations to identify boundary PEs. If
a PE is in the right most boundary of the robot(on its row) it
sends its column index to the leftmost PE of the
N N RMESH on that row. This is done in Step2 and Step3.
Similar the left most boundary PEs of the robot send their
column indices to the leftmost PEs of the N N RMESH on
their rows. Step4 and Step5 fulfill this. Every leftmost PE then
calculates the run length of the robot on that row. The next
step is to rank the row strips of the robot starts from the top of
the strip to the bottom(Step7). Note this operation is a special
case of the general rank operation. Here the PEs involved in
the ranking is in consecutive top to bottom fashion. Therefore
the ranking operation can be done in O(1) time by first identify
the top boundary PE(note the rank of this PE is 0). Followed
by one broadcasting of the row index and simple algebra, other
PEs can then determine their ranks. At this time the PEs that
Fig. 14. A WBP convex robot with two base points

7

are in the leftmost column of the N N RMESH have the run
length information of the robot on that particular row and the
ranking information. These information will then broadcast to
other planes by using Z bus. On receiving the run length
information at Step9 and rank at Step11, the PEs can then
compare the rank value with its k index. If these two values
match then the PE will broadcast the runlength information to
the PEs on its plane and this is done in Step12, 13 and 14 by
using plane bus. It is easily seen the complexity is O(1).


Step0 Tiling;
Step1 compute vertical-boundary(i,j,k);
form column bus;
Step2 If vertical-boundary(i,j,k) then
disconnect N switch;
broadcast(i);
temp(i,j,k) = content(bus);
If not obstacle(i,j,k) then
distance(i,j,k) =i-temp(i,j,k);
Step3 If vertical-boundary(i,j,k) then
DrawSegment(height(i,j,k)) toward south;
Step4 If A(i,j,k) then obstacle(i,j,k) = true;
Step5 form z bus;
Step6 If obstacle(i,j,k) then broadcast(distance(i,j,k))
temp(i,j,k)= content(bus)
if (k= = temp(i,j,k)) then broadcast(runLength(i,j,k));
runLength(i,j,k) = content(bus);
Step7 if obstacle(i,j,k) then
AdjacentUnionRight(runLength(i,j,k));

Fig. 16. Computing of configuration space for L-shaped robot

The algorithm to compute the configuration space obstacles
is shown in Figure 16. The algorithm assumes that the
digitized images of the obstacles and a robot are loaded into
plane zero of the RMESH computer. During the image
loading, two boolean variables, robot and obstacle, of each PE
are initialized. A PEs robot variable is initialized to true iff it
is a robot PE, i.e., it contains a pixel value of the robot. A PEs
obstacle variable is initialized to true iff it is a obstacle PE. It
is also assumed that the inversion operation has been
performed and resulted in the robot under discussion.
Like all the algorithms presented in the paper, algorithm of
Figure 16 is executed by every PE in the RMESH. Each
PE(i,j,k) has the following important variables that are related
to the current algorithm: robot(i,j,k), obstacle(i,j,k),
runLength(i,j,k), distance(i,j,k), height(i,j,k), vertical-
boundary(i,j,k), A(i,j,k), and temp(i,j,k). Each PE also has
three constants i,j,k, which form the ID of the PE. Hence, in
the algorithm symbols i,j,k refer to the constants i,j,k
respectively.
The temp variable is used for obtaining bus data by each PE
and does not have a significant role in the algorithm like the
rest of the variables. The obstacle and robot variables are
initialized during image loading as indicated earlier. In
addition, the loading phase also initializes each vertical-
boundary variable to false, A variable to zero, and distance
variable to zero.
Variable robot(i,j,k) is only used in Step0 by the tiling
procedure, which initializes the runlength(i,j,k) and
height(i,j,k) variables of each PE. After Step0, the height of
the robot is stored in the height(i,j,k) variable of each PE, and
the length of the i
th
horizontal segment is stored in the
runLength(i,j,k) variable of each PE in plane i. Thus, the
values of runLength(i,j,k) variables of the PEs belonging to the
same plane are the same.
In Step1, each obstacle PE checks its neighbor PEs to see if
it needs to assign true to its vertical-boundary(i,j,k) variable.
The shift operations are used for getting the values of the
obstacle(i,j,k) variables of the neighboring PEs. Once this is
done, the PEs set up the column buses for the next step. Step2
computes the values for the distance(i,j,k) variables. Each
boundary PE sets up its column sub bus and sends its row
index to the PEs down the south. Then each PE gets the row
index from the bus and determines its distance to the boundary
PE. The value of distance(i,j,k) will not be used later unless
PE(i,j,k) is or becomes an obstacle PE. Step 3 carries out the
vertical expansion. During the expansion, the value of A(i,j,k)
will be set by DrawSegment operation if PE(i,j,k) is on the
expansion path, i.e., PE(i,j,k) is a new obstacle PE. Step 4
reflects this fact by adjusting the obstacle variables. Step 5
prepares the Z buses so that the obstacle PEs can obtain their
horizontal expansion length. Getting the length is done in Step
6. Step 7 carries out the horizontal expansion and completes
the algorithm.

IV. CONCLUSIONS
Basic data manipulation operations on RMESH such as
odd-even phase shifting operation, DrawSegment operation,
AdjacentUnion operation, Image inversion operation were
conceptualized and their implementation were developed.
These operations may be used as basic building blocks to
develop algorithms to solve more complex problems
efficiently, which was demonstrated in this paper. Using these
operations along with other existing operations a novel
algorithm for computing configuration space obstacles was
developed.
The algorithm we developed for computing the
configuration space obstacles is for convex robot by using
D N N reconfigurable mesh with buses (RMESH), where
D is the diameter of the robot under consideration. The
algorithm is asymptotically optimal when time complexity is
concerned. The algorithm runs in constant time and uses
constant space. There are other interesting questions which we
did not address in this report. Can we reduce the size of the
RMESH and achieve the same optimal complexity? Can we
compute configuration space obstacles when arbitrary shape
robot is concerned? If yes, can we still achieve the constant
time and space complexity?

8

REFERENCES
[1] S. Bandi, D. Thalmann, A Configuration Space Approach for Efficient
Animation of Human Figures, IEEE Computer Society Workshop on
Motion of Non-Rigid and Articulated Objects, pp. 38 45, 1997.
[2] F. Dehne, A. Hassenklover, and J. Sack, "Computing the configuration
space for a robot on a mesh-of-processors, Proceedings 1989 ICPP.
vol. 3, pp. 40-47, 1989.
[3] Dinesh Manocha,Liangjun Zhang,Young J. Kim , C-DIST: efficient
distance computation for rigid and articulated models in configuration
space, Proceedings of the 2007 ACM symposium on Solid and
physical modeling, pp. 159-169, 2007.
[4] I. Ivanisevic, V. J. Lumelsky, "Configuration space as a means for
augmenting human performance in teleoperation tasks", IEEE
Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics",
vol. 30, no. 3, pp 471 484, 2000.
[5] J. Jenq and S. Sahni, Serial and Parallel Algorithms for the Medial
Axis Transform, IEEE Transactions on Pattern Analysis and Machine
Intelligence, Dec. pp 1218-1224, 1992.
[6] J. Jenq and S. Sahni, Reconfigurable Mesh Algorithms for
Fundamental Data Manipulation Operations in Computing on
Distributed memory Multiprocessors, NATO series F, ed. F. Ozguner,
Spring Verlag, pp 27-46, 1993.
[7] J. Jenq and W. Li, Optimal hypercube algorithms for robot
configuration space computation, Proceedings of the 1995 ACM
Symposium on Applied Computing, pp 182-186.
[8] J. Jenq and W. Li, Computing the Configuration Space for a Convex
Robot on Hypercube Multiprocessors, Proceedings of the 7th IEEE
Symposium of Parallel and Distributed Processing, pp 160-167, 1995.
[9] L. Kavraki, "Computation of Configuration-Space Obstacles Using the
Fast Fourier Trnasform", IEEE Transactions on Robotics and
Automation, vol. 11(3), pp 408-413, 1995.
[10] C.L. Lia,1, K.W. Chanb, S.T. Tanb, "A configuration space approach to
the automatic design of multiple-state mechanical devices", Computer-
Aided Design 31, pp 621653, 1999, Elsevier.
[11] T. Lozano-Perez and M. A. Wesley, "An algorithm for planning
collision-free paths among polyhedral obstacles," CACM, pp. 5609.-
570, 197.
[12] T. Lozano-Perez, "Spatial planning: A configuration space approach,"
IEEE Trans. on Computers, pp. 108-120, 1983..
[13] T. C. Manjunath, Gopala, Ashok Kusagur, B. G. Nagaraja , Simulation
& Implementation of Shortest Path Algorithm with a Mobile Robot
Using Configuration Space Approach, International Conference on
Advanced Computer Control, pp. 197-201, 2009.
[14] R. Miller, V. Prasanna-Kumar, D. Reisis, and Q. Stout, Parallel
Computations on Reconfigurable Meshes, IEEE Transactions on
Computers, vol. 42(6), pp 678-692, 1993.
[15] Andrzej Wytyczak-Partyka, Jerzy W. Rozenblit, Chuan Feng, Allan J.
Hamilton, Defining Spatial Regions in Computer-Assisted
Laparoscopic Surgical Training, IEEE International Conference on the
Engineering of Computer-Based Systems, pp. 176-183, 2009.
[16] P. Tzionas, A. Thanailakis, and P. Tsalides, "Collision-Free Path
Planning for a Diamond-Shaped Robot Using Two dimensional Cellular
Automata", IEEE Transactions on Robotics and Automation, vol.13(2),
pp 237-250, 1997.
[17] Wang Yuquan, Zhu Qidan, Zhou Fang, Wang Tong, "Path Planning for
Multi-Joint Manipulator Based on the Decomposition of Configuration
Space", International Conference on Intelligent Computation
Technology and Automation, pp. 661-664, 2009.


John Jenq is an associate professor of Computer Science Department at
Montclair State University, Montclair New Jersey. Dr. Jenq received his
Master of Science and PhD from University of Minnesota, Minneapolis in
1986 and 1991 respectively. His research interests include parallel and
distributed computation, image processing, pattern recognition, data mining,
algorithmic robotics, and internet applications . Dr. Jenq is a member of both
ACM and IEEE.

Dajin Wang is a professor of Computer Science Department at Montclair
State University. Dr. Wang received his B. Eng degree from Shanghai
University of Science and Technology in 1982, Master of Science and PhD
from Stevens Institute of Technology in 1986 and 1990 respectively. His
research interests include interconnection networks, fault tolerant computing,
parallel and distributed computing, wireless mobile and sensor networks, and
algorithmic robotics.

Wingning Li is a professor with Department of Computer Science and
Computer Engineering, University of Arkansas, where he has been serving
from 1998-present. Dr. Li obtained his B.S. Degree in Computer Science
University of Iowa, December 1982, his M.S. Degree in Computer Science
University of Minnesota, November 1985, and his Ph.D. Degree in Computer
Science University of Minnesata, September 1989. Dr. Li research interests
are in the areas of Computer-aided design for VLSI circuits, combinatorial
optimization, design and analysis of algorithms in both theoretical and
experimental settings, parallel computing, software reuse and construction,
and GUI design and development. Dr. Li is member of both ACM and IEEE.